Tuesday, April 22, 2008

Reply – Blog Comment 4-21-08

Fred Millar had a comment on one of the incident reviews from yesterday’s blog (see "Chemical Incident Review – 4-21-08"). He said: "Seems too difficult for the chemical industry to learn what the gasoline station folks did many years ago -- make nozzles and tank openings different/incompatible for incompatible chemicals."

Actually the problem is a little more complex than nozzles in holes when your doing bulk unloading for up to 60 different chemicals. Trucks and rail cars come with standard fittings for 2", 3" or 4" hoses so you are limited in available combinations.

In the last facility that I worked at we came up with a fairly good system that worked for 16 years without incident that I know of. We had three dedicated bulk-unloaders, no one else was authorized to unload from a tank wagon or rail car into a storage tank. This helped to limit the problem.

All of the hose connections to storage tanks were locked with two padlocks and each lock of the pair was keyed differently. The shift supervisor had a master key to one of the locks and QA had a keys to one lock on each valve. At the start of each shift the shift supervisor removed his lock from each line that was scheduled to deliver that day. To get the key from QA the bulk-unloader had to take the paperwork from the load that identified the chemical being delivered. Once QA approved the incoming material (usually by checking CoA values against the approved values) the technician gave the unloader the key to the appropriate storage tank.

This works well if you trust the paperwork that accompanies the truck. Like I say we never had a problem in 16 years. Of course, we never had a terrorist attack.

I’ve brainstormed this problem with a couple of different people over the last couple of years. The best solution to prevent a terrorist attack via this method is to do the reactivity test I briefly described in the blog. You take a small sample (a couple of drops) of the incoming chemical and place in a small (about 1 oz/30 mL) sample from the storage tank and observe for a reaction. Ideally this should be done in a fume hood behind a small blast shield, just in case. This does not stop all quality problems, but it does take care of the catastrophic ones.

Everyone that has worked in chemical manufacturing has heard of or experienced at least one wrong chemical into the wrong tank story. Almost everyone that I have heard was due to a truck driver being required to do the unloading. It is amazing how many facilities still allow/require that. A company that I worked for required the receiving facility manager to sign a blanket release before we would allow one of our truck drivers to unload at their facility. Still about once a year one of our drivers would unload into the wrong tank.

No comments:

/* Use this with templates/template-twocol.html */